Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Operator inference”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 433 records · Page 24
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Probabilistic Modeling of Commercial Building Occupancy Patterns Using Location-Based Map Data: Preprint

Considering occupancy patterns is crucial to simulate buildings' energy use. Current energy models use inputs that simplify the actual diversity in occupancy into static occupancy patterns and are not able to represent the numerous variations in occupancy patterns between buildings and across different locations. Recently, inferring occupancy schedules from metered electricity consumption data was used to model occupancy in commercial buildings. However, the translation from metered data to occupancy schedules requires many assumptions that might not capture the reality, and the process is hindered by the availability of data from advanced metering infrastructure. With the development of information technologies, occupancy modeling should not be limited to traditional approaches. The prevalence of social networks and location services with real-time user feedback provides publicly accessible data via Maps Application Programming Interfaces (APIs) such as Google Maps, SafeGraph, Mapbox, Foursquare, etc. This paper presents an automated framework for modeling parametric occupancy patterns using such APIs to calibrate commercial district buildings' energy models. This process includes three main steps: data extraction and processing, parametric schedules generation, and schedules integration. We demonstrated this framework in districts where we used maps API to generate more accurate behavioral patterns for operations and electric vehicle charging events. We used these patterns to determine differences in energy use across key sociodemographic and spatial parameters. The presented method has the potential for worldwide applications. Users can utilize this framework to extract data for selected locations of interest to create more realistic behavioral patterns for commercial facilities across different districts.

building energy modeling↗
DOE OSTI2021

Classification and computation of extreme events in turbulent combustion

In the design of practical combustion systems, ensuring safety and reliability is an important requirement. For instance, reliably avoiding lean blowout, flame flashback or inlet unstart is critical for ensuring safe operation. Currently, the science of predicting such events is based on prior experience, limited modeling or diagnostic tools and purely statistical approaches. Even though computational and experimental tools for studying combustion devices have vastly advanced in the last three decades, the analysis of such failure events has not been pursued widely. While the use of data for model development and calibration is being widely accepted, the extension to failure events introduces numerous challenges. In particular, the focus here is on so-called data-poor problems, where the cost of generating data is extremely high and is not easily amenable to existing computational and experimental approaches. Data-poor problems are particularly relevant when related to extreme events (also called anomalous events) that can lead to catastrophic failure of the system. It is argued that transient events that describe such failure can have different causal mechanisms. To develop the scientific inference process, a classification of such problems is used to determine specific modeling paths as well as computational tools needed. Research opportunities in the emerging field of extreme event prediction are highlighted in order to identify critical and immediate needs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

In-Situ TEM Molten Salt Corrosion

Molten salt reactors (MSRs) offer a compelling pathway for next-generation nuclear energy, with advantages in thermal efficiency, inherent safety, and flexible fuel management. Yet, halide-based molten salts introduce significant materials challenges, particularly alloy corrosion. Alloy performance in these environments ultimately depends on understanding how corrosion initiates and progresses at the nanoscale, however most existing models rely on post-exposure characterization, leaving degradation mechanisms largely inferred rather than directly observed. NiCr alloys have garnered interest in MSRs applications as the Ni-based matrix provides strength and creep resistance, while Cr content offers oxidation resistance in air. However, NiCr corrosion resistance in chloride salts has proven poor due to preferential chromium dissolution, the formation of Cr-depleted pathways, and grain-boundary attack. This work aims to directly visualize corrosion of Ni-20Cr exposed to LiCl-KCl using in-situ Transmission Electron Microscopy (TEM) to capture real-time microstructural evolution during corrosion. Experiments will be performed at ~800 °C under controlled pressure conditions while utilizing Energy-Dispersive X-ray Spectroscopy (EDS) to analyze elemental redistribution. Observation of chromium depletion fronts, associated surface restructuring, and localized chloride enrichment are expected. Ultimately, this study is expected to provide a link between microscale processes and the macroscopic degradation behaviors relevant to MSR operation in advanced reactor environments. Simultaneously, this approach enables future in-situ investigations regarding alloy composition, salt chemistry, and their influence on corrosion pathways and long-term stability.?

36 - MATERIALS SCIENCE↗
DOE OSTI2024

Revealing Decision Conservativeness Through Inverse Distributionally Robust Optimization

This paper introduces Inverse Distributionally Robust Optimization (I-DRO) as a method to infer the conservativeness level of a decision-maker, represented by the size of a Wasserstein metric-based ambiguity set, from the optimal decisions made using Forward Distributionally Robust Optimization (F-DRO). By leveraging the Karush-Kuhn-Tucker (KKT) conditions of the convex F-DRO model, we formulate I-DRO as a bi-linear program, which can be solved using off-the-shelf optimization solvers. Additionally, this formulation exhibits several advantageous properties. We demonstrate that I-DRO not only guarantees the existence and uniqueness of an optimal solution but also establishes the necessary and sufficient conditions for this optimal solution to accurately match the actual conservativeness level in F-DRO. Furthermore, we identify three extreme scenarios that may impact I-DRO effectiveness. Our case study applies F-DRO for power system scheduling under uncertainty and employs I-DRO to recover the conservativeness level of system operators. Numerical experiments based on an IEEE 5-bus system and a realistic NYISO 11-zone system demonstrate I-DRO performance in both normal and extreme scenarios. An extended version of this paper with additional analyses is available at li2024revealing.

distributionally robust optimization↗
DOE OSTI2024

Homomorphic Encryption for Electrical Metering Aggregation: Protecting the Privacy of Building Tenants

Electrical meters are devices that measure consumer electricity usage. The data collected by these meters is necessary for utility billing and electrical grid management but can also be used to assess the environmental impact of buildings. Prior research has found that unprotected metering data could potentially be used to infer some information about the behaviors of building tenants by detecting changes in electricity usage. For example, a period of low electricity usage could suggest that the tenants are not in the building. As smart metering becomes more common, there is a growing need for data privacy protections for metering data that do not negatively impact the quality and availability of data used for energy management and billing applications. To identify potential solutions, we developed a Python-based data aggregation platform to analyze the potential efficacy of privacy-enhancing technologies for energy metering applications. This platform aggregates groups of metering sites into virtual buildings, which could potentially detach changes in electrical activity from individual tenants, making it more difficult to track the activity of a specific tenant. To further protect data during analysis, this project utilizes homomorphic encryption as part of its initial approach. Homomorphic encryption offers a means of protecting energy consumption data while permitting mathematical operations to be performed without the need to know the data contents. This allows for data to be processed into usable statistics without revealing energy consumption information. A series of homomorphic encryption libraries were evaluated to determine their applicability and limitations in the context of metering data. The use of these techniques may help to reassure consumers and encourage further adoption of smart grid infrastructure.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

On the impact of tidal generation and energy storage integration in PV-rich electric distribution systems

Deep decarbonization of power system operations requires the maximal utilization of available renewable resources. At distribution-level operations, however, grid operators can face numerous challenges in integrating renewables at scale owing to the inherent intermittence of renewable energy resources. These include phenomena such as voltage fluctuations, which are typically mitigated through control actuators such as on-load tap changers (OLTC) as well as energy storage devices, such as battery energy storage systems (BESS). On the one hand, high intermittence of the available renewable portfolio may require increasingly aggressive control of actuators, thereby accelerating the probability of equipment failure. On the other hand, integrating BESS operations and having a diverse renewable generation portfolio can typically help stagger power/energy flow to mitigate the aforementioned adverse impacts. In this paper, we employ a Bayesian framework for equipment lifetime estimation to understand the impact of including tidal energy resources and BESS in distribution system operations for feeders having substantial distribution photovoltaic generation. Our results indicate that while tidal energy alone may slightly decrease equipment reliability, the adverse impact on reliability is significantly magnified by a generation portfolio consisting of tidal generation and photovoltaic generation. Here, we also study the tidal and photovoltaic hosting capacity problem with and without energy storage systems using equipment reliability as an added constraint. We conclude that energy storage increases the reliability-constrained hosting capacity of the distribution system.

14 SOLAR ENERGY↗
DOE OSTI2023

Bayesian uncertainty quantification of tristructural isotropic particle fuel silver release: Decomposing model inadequacy plus experimental noise and parametric uncertainties

Tristructural isotropic (TRISO) particle fuel is one of the most promising fuel concepts enabling high temperature and high burnup reactor operation. One dominant source of radioactivity released from the TRISO particles is silver (Ag), which is subject to a high release fraction and long decay life compared to other fission products. Previous modeling efforts using the fuel performance code BISON indicated nonnegligible uncertainties in modeling the diffusion process of fission products in TRISO compared to the Advanced Gas Reactor experiments. The overall uncertainties observed when modeling the fission product diffusion can result from uncertainties in model parameters, noisy experimental measurements, and deficiencies in the developed models. The three types of underlying uncertainties have not yet been properly quantified in open literature. Here, this paper presents the Bayesian uncertainty quantification (UQ) using massively parallelizable Markov chain Monte Carlo samplers. The uncertainties due to model parameters, model inadequacy, and experimental measurement noise are quantified, with the σ term used to represent the sum of the model inadequacy and measurement noise uncertainties. It is worth noting that this is the first time the σ term is inferred for nuclear fuel experiments, as compared to using prescribed values for uncertainty quantification in previous work. The parallelizable Markov chain Monte Carlo samplers efficiently infer the model parameters and the σ term, giving insight into physical parameters like diffusion coefficients and the combined model discrepancy and measurement noise. A subsequent forward uncertainty quantification (UQ) is also performed based on the calibration results to generate more accurate predictions of the Ag release. The model inadequacy plus experimental noise is the most dominant source of uncertainty compared to the parametric uncertainty. All the UQ analyses presented in this work are based on the second series of the irradiation experiments in the Advanced Gas Reactor program.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

Survey of tungsten gross erosion from main plasma facing components in WEST during a L-mode high fluence campaign

An initial high fluence campaign was performed in WEST, in 2023, on the newly installed actively cooled tungsten divertor composed of ITER-grade monoblocks. The campaign consisted in the repetition of a 60 s long Deuterium L-mode pulse in attached divertor conditions, cumulating over 10000s of plasma exposure. A maximum deuterium fluence of approximately 5⋅1⁢026 m−2 was reached in the outer strike point region, representative of a few high performance ITER pulses. Gross tungsten erosion inferred from visible spectroscopy shows that the most eroded plasma facing component is the inner divertor target with rates ten times larger than on the outer divertor target. The outer midplane tungsten bumpers, located a few centimeters from the plasma, show gross erosion rates two times lower than at the outer divertor. We conclude that the outer midplane bumpers have a negligible contribution to the long range tungsten migration and deposition onto the lower divertor. The cumulated gross erosion rate on the inner divertor translates in an effective gross erosion thickness of about 20μ⁢m, while it is about 2μ⁢m for the outer divertor. Strikingly, these orderings coincide with the thickness of deposits found locally on the divertor: the exposed surfaces of high field side monoblocks are covered with several tens of μ⁢m tungsten deposits, while on the lower field side, few μ⁢m thin tungsten deposits are only found on the magnetically shadowed parts of monoblocks. The strong impact of those deposits on WEST operation, namely perturbation of surface temperature measurement with infra-red thermography, and the emission of flakes causing radiative perturbation of the confined plasma, calls for anticipating similar issues in ITER. In particular, the start of research operation shall consider the definition of a divertor erosion budget in order to anticipate the formation of deleterious deposits.

Fedorczak, N.↗
DOE OSTI2023

Isotope effects on energy transport in the core of ASDEX-Upgrade tokamak plasmas: Turbulence measurements and model validation

Design and operation of future tokamak fusion reactors using a deuterium–tritium 50:50 mix requires a solid understanding of how energy confinement properties change with ion mass. This study looks at how turbulence and energy transport change in L-mode plasmas in the ASDEX Upgrade tokamak when changing ion species between hydrogen and deuterium. For this purpose, both experimental turbulence measurements and modeling are employed. Local measurements of ion-scale (with wavevector of fluctuations perpendicular to the B-field k⊥< 2 cm−1, k⊥ρs< 0.2, where ρs is the ion sound Larmor radius using the deuterium ion mass) electron temperature fluctuations have been performed in the outer core (normalized toroidal flux ρTor=0.65−0.8) using a multi-channel correlation electron cyclotron emission diagnostic. Lower root mean square perpendicular fluctuation amplitudes and radial correlation lengths have been measured in hydrogen vs deuterium. Measurements of the cross-phase angle between a normal-incidence reflectometer and an ECE signal were made to infer the cross-phase angle between density and temperature fluctuations. The magnitude of the cross-phase angle was found larger (more out-of-phase) in hydrogen than in deuterium. TRANSP power balance simulations show a larger ion heat flux in hydrogen where the electron-ion heat exchange term is found to play an important role. These experimental observations were used as the basis of a validation study of both quasilinear gyrofluid trapped gyro-Landau fluid-SAT2 and nonlinear gyrokinetic GENE codes. Linear solvers indicate that, at long wavelengths (k⊥ρs<1), energy transport in the deuterium discharge is dominated by a mixed ion-temperature-gradient (ITG) and trapped-electron mode turbulence while in hydrogen transport is exclusively and more strongly driven by ITG turbulence. The Ricci validation metric has been used to quantify the agreement between experiments and simulations taking into account both experimental and simulation uncertainties as well as four different observables across different levels of the primacy hierarchy.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Analysis of Polarized Dust Emission Using Data from the First Flight of SPIDER

Using data from the first flight of Spider and from the Planck High Frequency Instrument, we probe the properties of polarized emission from interstellar dust in the Spider observing region. Component-separation algorithms operating in both the spatial and harmonic domains are applied to probe their consistency and to quantify modeling errors associated with their assumptions. Analyses of diffuse Galactic dust emission spanning the full Spider region demonstrate (i) a spectral energy distribution that is broadly consistent with a modified-blackbody (MBB) model with a spectral index of β d = 1.45 ± 0.05 (1.47 ± 0.06) for E (B)-mode polarization, slightly lower than that reported by Planck for the full sky; (ii) an angular power spectrum broadly consistent with a power law; and (iii) no significant detection of line-of-sight polarization decorrelation. Tests of several modeling uncertainties find only a modest impact (~10% in σ r ) on Spider's sensitivity to the cosmological tensor-to-scalar ratio. The size of the Spider region further allows for a statistically meaningful analysis of the variation in foreground properties within it. Assuming a fixed dust temperature T d = 19.6 K, an analysis of two independent subregions of that field results in inferred values of β d = 1.52 ± 0.06 and β d = 1.09 ± 0.09, which are inconsistent at the 3.9σ level. Furthermore, a joint analysis of Spider and Planck 217 and 353 GHz data within one subregion is inconsistent with a simple MBB at more than 3σ, assuming a common morphology of polarized dust emission over the full range of frequencies. This evidence of variation may inform the component-separation approaches of future cosmic microwave background polarization experiments.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

High-volume and -adiabat capsule (“HVAC”) ignition: Lowered fuel compression requirements using advanced Hohlraums

Lower-than-expected deuterium–tritium fuel areal densities have been experimentally inferred across a variety of high-convergence, nominally low-adiabat implosion campaigns at the National Ignition Facility (NIF) using cylinder-shaped Hohlraums [Hurricane et al., Phys. Plasmas 26, 052704 (2019)]. A leading candidate explanation is the presence of atomic mix between the fuel and ablator from hydrodynamic instability growth [Clark et al., Phys. Plasmas 26, 050601 (2019)], leading to reduced fuel compressibility and an effectively higher (in-flight) fuel adiabat α. Tolerating a high-α implosion can be obtained with significantly higher capsule absorbed energy Ecap according to the one-dimensional (1-D) ignition-threshold-factor analytic scaling [S. Atzeni and J. Meyer-ter-Vehn, Nucl. Fusion 41, 465 (2001)], ITF∼Ecap·α−1.8. Recent experiments with large Al shells in rugby-shaped Hohlraums have established high laser-capsule coupling efficiencies of ≽ 30% [Ping et al., Nat. Phys. 15, 138 (2019)], enabling a path to Ecap≽ 0.5 MJ at the NIF and increased performance margin M ≡ ITF − 1. The ability to operate at high adiabat with large capsules using nonstandard Hohlraums leads to the predicted onset of a volume-ignition mode, defined as when both the entire fuel is the “hot spot” and inertial confinement is principally provided by the ablator compared with the compressed fuel. Such an ignition mode, normally reserved for high-Z targets, e.g., double shells [Amendt et al., Phys. Plasmas 14, 056312 (2007)], is predicted to lead to lower fuel convergence and less exposure to mix due to the intended high adiabat—but at the expense of ∼3–4 × reduced (1-D) yield compared with conventional central hot-spot ignition designs.

Amendt, Peter (ORCID:0000000174021777)↗
DOE OSTI2021

The scalar chemical potential in cosmological collider physics

Non-analyticity in co-moving momenta within the non-Gaussian bispectrum is a distinctive sign of on-shell particle production during inflation, presenting a unique opportunity for the “direct detection” of particles with masses as large as the inflationary Hubble scale ( H ). However, the strength of such non-analyticity ordinarily drops exponentially by a Boltzmann-like factor as masses exceed H . In this paper, we study an exception provided by a dimension-5 derivative coupling of the inflaton to heavy-particle currents, applying it specifically to the case of two real scalars. The operator has a “chemical potential” form, which harnesses the large kinetic energy scale of the inflaton, \( {\overset{\cdot }{\phi}}_0^{1/2}\approx 60H \) , to act as an efficient source of scalar particle production. Derivative couplings of inflaton ensure radiative stability of the slow-roll potential, which in turn maintains (approximate) scale-invariance of the inflationary correlations. We show that a signal not suffering Boltzmann suppression can be obtained in the bispectrum with strength f NL ~ \( \mathcal{O} \) (0 . 01–10) for an extended range of scalar masses \( \lesssim {\overset{\cdot }{\phi}}_0^{1/2} \) , potentially as high as 10 15 GeV, within the sensitivity of upcoming LSS and more futuristic 21-cm experiments. The mechanism does not invoke any particular fine-tuning of parameters or breakdown of perturbation-theoretic control. The leading contribution appears at tree-level , which makes the calculation analytically tractable and removes the loop-suppression as compared to earlier chemical potential studies of non-zero spins. The steady particle production allows us to infer the effective mass of the heavy particles and the chemical potential from the variation in bispectrum oscillations as a function of co-moving momenta. Our analysis sets the stage for generalization to heavy bosons with non-zero spin.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

DETECTING FIRE WITH MACHINE LEARNING-ENABLED VISUAL MONITORING FOR NUCLEAR POWER PLANT ENVIRONMENTS

Nuclear power plants are experiencing significant cost challenges to remain competitive with other energy-generation utilities. Unlike other industries, the cost of operation and maintenance activities is mostly attributed to workforce costs. To mitigate this, nuclear power plant stakeholders are increasingly interested in the development and deployment of machine learning methods to potentially automate or augment manually intensive tasks to reduce costs, especially for monitoring activities. One monitoring function that is visually demanding and that can occur frequently to meet the requirements of a fire protection program is visually monitoring an area for fire occurrence. Currently, fire watch activities consist of a worker physically stationed at a given location with the sole responsibility of observing a given area to ensure a fire is detected and mitigated promptly. This effort focused on the development and evaluation of a suitable deep convolutional neural network to classify individual video frames at a sub-second frequency for the occurrence of “fire” and “no fire” in varying industrial environments similar to nuclear power plants. It is believed that a trained neural network model could be integrated with existing facility video surveillance camera feeds to generate alerts when fire inferences occur in individual frames captured at sub-second temporal resolutions. Extensive effort was dedicated to identifying and curating suitable imagery training data representing varying environments and scene settings with and without flame features to maximize generalization in nuclear power plant environments. The data collection effort resulted in the aggregation of a large, labeled image library exceeding 12,000 images to support model training for diverse industrial environments. A deep neural network model incorporating parallel multi-scale capabilities was developed and trained to support accurate image-based detection of flame incidents of varying sizes and spectral feature properties within heterogeneous scenes. Analysis results show that the trained model can achieve high inference accuracy despite heterogeneous scene environments and components. Testing accuracy exceeded 95.0 percent with very low false positive and false negative inferences.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2021

Characterization of Self-Magnetic Pinch (SMP) radiographic diode performance on RITS-6 at Sandia National Laboratories: 1) Diode Dynamics, DC Heating to extend Radiation Pulse

Radiographic diodes focus an intense electron beam to a small spot size to minimize the source area of energetic photons for radiographic interrogation. The self-magnetic pinch (SMP) diode has been developed as such a source and operated as a load for the RITS-6 Inductive Voltage Adder (IVA) driver. While experiments support the generally accepted conclusion that a 1:1 aspect diode (cathode diameter equals anode-cathode gap) delivers optimum SMP performance, such experiments also show that reducing the cathode diameter, while reducing spot size, also results in reduced radiation dose, by as much as 50%, and degraded shot reproducibility. Analyzation of the effective electron impingement angle on the anode converter with time made possible by a newly developed dose-rate array diagnostic indicates that fast-developing oscillations of the angle are correlated with early termination of the radiation pulse on many of the smaller-diameter SMP shots. This behavior as a function of relative cathode size persists through experiments with output voltages and currents up to 11.5 MV and 225 kA, respectively, and with spot sizes below ~ few mm. Since simulations to date have not predicted such oscillatory behavior, considerable discussion of the angle-behavior of SMP shots is made to lend credence to the inference. There is clear anecdotal evidence that DC heating of the SMP diode region leads to stabilization of this oscillatory behavior. This is the first of two papers on the performance of the SMP diode on the RITS-6 accelerator.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

Recurrent Convolutional Deep Neural Networks for Modeling Time-Resolved Wildfire Spread Behavior

The increasing incidence and severity of wildfires underscores the necessity of accurately predicting their behavior. While high-fidelity models derived from first principles offer physical accuracy, they are too computationally expensive for use in real-time fire response. Low-fidelity models sacrifice some physical accuracy and generalizability via the integration of empirical measurements, but enable real-time simulations for operational use in fire response. Machine learning techniques have demonstrated the ability to bridge these objectives by learning first-principles physics while achieving computational speedups. While deep learning approaches have demonstrated the ability to predict wildfire propagation over large time periods, time-resolved fire-spread predictions are needed for active fire management. Here, in this work, we evaluate the ability of deep learning approaches in accurately modeling the time-resolved dynamics of wildfires. We use an autoregressive process in which a convolutional recurrent deep learning model makes predictions that propagate a wildfire over 15 min increments. We apply the model to four simulated datasets of increasing complexity, containing both field fires with homogeneous fuel distribution as well as real-world topologies sampled from the California region of the United States. We show that even after 100 autoregressive predictions representing more than 24 h of simulated fire spread, the resulting models generate stable and realistic propagation dynamics, achieving a Jaccard score between 0.89 and 0.94 when predicting the resulting fire scar. The inference time of the deep learning models are examined and compared, and directions for future work are discussed.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Enhancing Automotive Intrusion Detection Through Multi-Modal Fusion: A CAN FD-LiDAR Approach

As vehicles become smarter and more autonomous, they increasingly depend on advanced sensors and communication technologies to operate securely. However, such growing dependence on technology—whether it’s CAN (Controller Area Network) for internal communication or LiDAR (Light Detection and Ranging) for sensing the world around them—also expands the attack surface for the types of cyber attacks. Traditional intrusion detection systems (IDS) typically monitor these systems in isolation, limiting their ability to detect sophisticated, crosssystem attacks. To address this, we propose a multi-modal fusion approach that combines real-world CAN FD signals (from the HCRL dataset) with LiDAR features (from the nuScenes dataset) to enhance attack detection. Our method employs a twostage ensemble approach. Calibrated XGBoost and LightGBM models initially process CAN FD (Fuzzing Data) and LiDAR data independently, detecting timing anomalies and space abnormalities. They are subsequently logarithmically combined with a logistic regression meta-model along with 17 engineered features capturing cross-modal behavior, prediction conflicts, and nonlinear interactions. This approach achieves an AUC of 0.87 and an F1-score of 0.82, surpassing single-modality baselines and early fusion methods, at merely 2 ms inference latency. Compared with deep learning competitors, it is 3 times more efficient, providing a lightweight, interpretable, and real time solution to automotive cybersecurity.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Sensor selection and tool wear prediction with data‐driven models for precision machining

Abstract Estimation of tool wear in precision machining is vital in the traditional subtractive machining industry to reduce processing cost, improve manufacturing efficiency and product quality. In this vein, fusion of time and frequency‐domain features of commonly sensed signals can provide an early indication of tool wear and improve its prediction accuracy for prognostics and health management. This paper presents a data‐driven methodology and a complete tool chain for the inference of precision machining tool wear from fused machine measurements, such as cutting force, power, audio and vibration signals, and quantify the usefulness of each measurement. Indicators of tool wear are extracted from time‐domain signal statistics, frequency‐domain analysis, and time‐frequency domain analysis. Correlation coefficients between the extracted features (indicators) and the tool wear are used to select the most informative features. Principal Component Analysis and Partial Least‐Squares are used to reduce the dimensionality of the feature space. Regression models, including linear regression, support vector regression, Decision tree regression, neural network regression and Gaussian process regression, are used to predict the tool wear using data from a Haas milling machine performing spiral boss face milling. The performance of the regression models based on subsets of sensors validates the preliminary estimates about the saliency of the sensors. The experimental results show that the proposed methods can predict the machine tool wear precisely, with readily available sensor measurements. Neural network and Gaussian process regression were able to achieve good estimates of tool wear at different machine operating conditions. The most informative signal in predicting tool wear was shown to be the vibration signal. Time‐frequency domain features were the most informative features among the combination of features of three domains. In addition, using partial least squares components extracted from the original features of signals led to higher prediction accuracy.

Han, Seulki↗
DOE OSTI2021

Benchmarking the Performance of Neuromorphic and Spiking Neural Network Simulators

Software simulators play a critical role in the development of new algorithms and system architectures in any field of engineering. Neuromorphic computing, which has shown potential in building brain-inspired energy-efficient hardware, suffers a slow-down in the development cycle due to a lack of flexible and easy-to-use simulators of either neuromorphic hardware itself or of spiking neural networks (SNNs), the type of neural network computation executed on most neuromorphic systems. While there are several openly available neuromorphic or SNN simulation packages developed by a variety of research groups, they have mostly targeted computational neuroscience simulations, and only a few have targeted small-scale machine learning tasks with SNNs. Evaluations or comparisons of these simulators have often targeted computational neuroscience-style workloads. In this work, we seek to evaluate the performance of several publicly available SNN simulators with respect to non-computational neuroscience workloads, in terms of speed, flexibility, and scalability. We evaluate the performance of the NEST, Brian2, Brian2GeNN, BindsNET and Nengo packages under a common front-end neuromorphic framework. Our evaluation tasks include a variety of different network architectures and workload types to mimic the computation common in different algorithms, including feed-forward network inference, genetic algorithms, and reservoir computing. We also study the scalability of each of these simulators when running on different computing hardware, from single core CPU workstations to multi-node supercomputers. Our results show that the BindsNET simulator has the best speed and scalability for most of the SNN workloads (sparse, dense, and layered SNN architectures) on a single core CPU. However, when comparing the simulators leveraging the GPU capabilities, Brian2GeNN outperforms the others for these workloads in terms of scalability. NEST performs the best for small sparse networks and is also the most flexible simulator in terms of reconfiguration capability NEST shows a speedup of at least 2x compared to the other packages when running evolutionary algorithms for SNNs. The multi-node and multi-thread capabilities of NEST show at least 2x speedup compared to the rest of the simulators (single core CPU or GPU based simulators) for large and sparse networks. We conclude our work by providing a set of recommendations on the suitability of employing these simulators for different tasks and scales of operations. We also present the characteristics for a future generic ideal SNN simulator for different neuromorphic computing workloads.

97 MATHEMATICS AND COMPUTING↗